1 Suicide and Primary Care Older decedents more likely overall to have visited primary care in previous month 58% of persons aged 55 years or older 23% of persons aged 35 and younger Limitation old data, study written in 2002 We screen for many other leading causes of death (colon cancer, hypertension), so WHY NOT screen systematically for suicidal risk?
2 Screening in Primary Care Most of primary care clinicians screening for suicidal ideation do so as part of a screen for major depression USPSTF recommends screening for major depression ONLY IF effective depression interventions with monitoring are available Depression is a risk factor for suicidal ideation, but not necessarily a leading risk factor for suicidal behavior
3 Current Approaches to Screening in Primary Care Lack of formal evidence-based recommendations explicit to suicidal ideation No gold standard screening instrument Q9 from the PHQ-9 Over the last 2 weeks, how often have you had thoughts you would be better off dead, or of hurting yourself in some way?
4 Challenges to Screening in Primary Care Lack of standardized screening instrument Limited research evidence that screening for suicidal ideation prevents suicide In some settings there is a lack of adequate interventions for high risk patients Persistent stigma that screening may induce suicidal ideation
5 Competing demands If you actually took the time to do all of the recommended preventive services with every patient and have all of the necessary conversations and address all of the concerns necessary, you would be working 18 hours a day -Agency for Health Research and Quality Clinical Advisor
6 Screening at Denver Health The Denver Health (DH) System of community clinics cares for about 30% of the population in Denver DH Primary Care does not screen all patients New systematic alcohol (CAGE) and drug abuse screen, but no SI screen Focused screening in the context of identified emotional/psychosocial distress
7 Behavioral Health Integration at Denver Health 5 of 9 community clinics have behavioral health clinicians (BHCs, Ph.D., Psy.D.) BHCs screen for suicidal ideation on all patients referred to them for emotional distress assessment of suicidal ideation versus nonsuicidal morbid ideation If suicidal ideation then assess risk and protective factors
8 Screening by Primary Care Clinicians Anecdotal information Some screening for patients with mental disorder, emotional distress, substance abuse Relatively clear protocol for the rare patient at high risk or with active ideation Contact on-call psychiatrist and/ or transfer patient to the psychiatric emergency room Place 72 hour mental health hold if necessary
9 Screening as part of a depression RCT at Denver Health 3027 different English-speaking persons aged 18+ contacted between 4/1/10 and 3/31/12 2 days prior to a primary care visit Invited to participate in a process to determine eligibility for a depression study Excluded immediately for current depression treatment, bipolar d/o, no phone, and lifethreatening physical comorbidity
10 Screening as part of a depression RCT at Denver Health 1161 screened with a 2-question instrument (PHQ-2) assessing anhedonia and depressed mood 443 scoring above threshold (2+ points out of 6, at least 2 points on either question) 389 took full PHQ-9 Q9 used as a screen for suicidal ideation
11 Screening as part of a depression RCT at Denver Health 97 (25% of 389) endorsed any 2-week ideation Similar to 33% with any SI, recent VA study (Yano EM 2012) Further assessment with 5-question MacArthur instrument
12 MacArthur Suicide Risk Assessment Questions 1. In the past month, have you made any plans or considered a method that you might use to harm yourself? 2. Have you ever attempted to harm yourself? 3. There s a big difference between having a though and acting on a thought. Do you think you might actually make an attempt to hurt yourself in the near future? 4. In the past month have you told anyone that you were going to commit suicide, or threatened that you might do it? 5. Do you think there is any risk that you might hurt yourself before you see your doctor the next time? If yes to any question then stratify to high risk.
13 Increasing Assessment for Suicidality in Primary Care? Most primary care clinicians do not routinely assess/ screen for suicidality (Graham 2011) On-site Mental Health services increase confidence to screen Female providers with lower confidence in assessing and treating suicidality Increasing confidence during residency for primary care clinicians may increase screening
14 Unanswered Questions Need better, more recent U.S. data on rates of contact with mental health and with primary care over months prior to death Need more data on mechanisms of action in contacts between health care providers and at-risk individuals, and on potential protective processes (Luoma 2002)
15 Suicide Screening in Emergency Departments Emmy Betz, MD, MPH Assistant Professor, Emergency Medicine University of Colorado School of Medicine
16 Joint Commission NPSG 15 NPSG Identify patients at risk for suicide Applies to psychiatric hospitals and patients being treated for emotional or behavioral disorders in general hospitals, including EDs Conduct a risk assessment that identifies specific patient characteristics and environmental features that may increase or decrease the risk for suicide. Address the patient s immediate safety needs and most appropriate setting for treatment. When a patient at risk for suicide leaves the care of the hospital, provide suicide prevention information (such as a crisis hotline) to the patient and his or her family. Slide 41
17 Current Local Practices As of March 19, 2012, all patients at the UCH ED screened Patient Safety Screener Question 1. Over the past 2 weeks, have you felt down, depressed, or hopeless? Yes No Refused Patient unable to complete 2. Over the past 2 weeks, have you had thoughts of killing yourself? Yes No Refused Patient unable to complete 3. Have you ever attempted to kill yourself? Yes No Refused Patient unable to complete 4. If Yes to item 3, ask: when did this last happen? Within the past 24 hours (including today) Within the last month (but not today) Between 1 and 6 months ago More than a six months ago Refused Patient unable to complete 5. Do you have a plan to hurt or kill yourself? Yes No Refused Patient unable to complete Physician notified Security notified Action Notify physician Ask question 4 Notify physician
18 Emergency Dept Safety and Follow-up Eval Study ED SAFE Aims: Test Universal Screening and Telephonic Intervention Figure 1: Overview of the phases and studies Screening Introduced Intervention Introduced Treatment As Usual Provide usual and customary screening and care Screening Alone Use Patient Safety Screener, sites handle positive screens per usual and customary care Brief ED Intervention (1) Question, Persuade, Refer (QPR) by primary nurse (2) Mental health evaluation (if appropriate) Intervention Safety Care-chain Post-ED Counseling (1) Coping Long-term with Attempted Suicide Program (CLASP-ED) (2) Up to 7 sessions with patient, 4 with significant other Screening Evaluation Primary Outcome (1) Rate of detection of ideation/behavior Secondary Outcomes (1) Receipt of a personalized safety plan (2) Behavioral health engagement (3) Suicide behavior Care-chain Evaluation Primary Outcome (1) Suicide behavior Secondary Outcomes (1) Receipt of a personalized safety plan (2) Behavioral health engagement Slide 43
19 Occult SI in % of ED Visits Single Sites, Various Indices of SI Study Definition of Ideation Freq (%) Boudreaux (2005) Boudreaux (2006) Boudreaux (2008) Claassen (2005) Ilgen (2009) Thoughts of, wanted to commit suicide 20/243 (8) Thoughts of, wanted to commit suicide 11/178 (6) Thoughts of, wanted to commit suicide 12/476 (3) Thoughts of death, better off dead 185/1590 (11.6) Thought about or wanted to kill self 134/1590 (8.4) Planning detected by providers 6/31 (19) Better off dead, hurting self (PHQ9, #9) 447/5641 (8) M. Allen
20 Occult SI in % of ED Visits Multicenter Study Definition of Ideation Freq (%) Allen (2010) CSSRS Passive SI CSSRS Active SI Any SI and history or attempt 79/1068 (7.5) 24/1068 (2.25) 12/1068 (3.3) ED-SAFE Retro Any mention of suicidal behavior 23 / 800 (2.9) ED-SAFE TAU Any mention of suicidal behavior 2771 / 94,385 (2.9) M. Allen
21 Current Practice ED-SAFE retrospective (n=800): 5% all ED patients screened 2.9% of all ED patients had SI or behavior Only 36% were evaluated by a mental health professional Patients screened: 90% reported psychiatric problem at triage 59% had current or past SI or behavior 21% reported self-harm behavior at triage 33-36% had documentation of alcohol or drug abuse 92% had 1+ characteristic suggesting suicide risk male, age 65+, psych complaint, history of substance misuse Ting et al. Multicenter Study of Predictors of Suicide Screening in Emergency Departments. Acad Emerg Med 2012.
22 Does screening identify high risk patients? ED-SAFE Treatment as usual (n=94,385) Screened: 26% of all ED patients screened SI or suicidal behavior: 2.9% same as retrospective! Local (UCH) changes with universal screening: Screened: Increased from 16% to 68% Documented self-harm: No change (also 3%!)
23 Care of Suicidal Patients in EDs Begins with assessment by ED physician If ED physician concerned, next step varies: On-site psychiatry or trained social workers (available 24/7 vs limited hours) versus Off-site psychiatry or mental health team This is more common in rural & smaller hospitals, and causes issues because of delays in patient care and concerns about long lengths of stay
24 Problems Limited inpatient beds and outpatient resources long ED waits for suicidal patients under lessthan-ideal circumstances Limited mental health professional availability Especially in rural areas or at smaller hospitals With time pressures and growing volume, these issues add to provider frustration
25 Problems Provider training ED physicians trained to ask about SI or suicidal behavior but not trained in risk assessment Traditionally have left further risk stratification up to mental health experts ED-SAFE Provider Survey (n=631) Confident in skills to screen for suicidal thoughts or behavior but not in skills for risk assessment, counseling, safety plans or finding referral resources Also concerns about staffing, support by leadership, and clinical priorities
26 Provider Self-Confidence in Skills for Care of ED Patients, By Provider Type (n=631) 100% Screen for Suicidality* Further Assess Suicide Risk** Provide Brief Counseling** Create Personalized Safety Plan*** Find Referral Resources 75% 50% 25% Disagree / Strongly disagree Uncertain Agree / Strongly agree 0% From ED-SAFE Provider Survey (phase 1); *P 0.05, **P 0.01, ***P under Pearson chi-square
27 Provider Opinions of Local ED Environment; by Provider Type (n=631) Hardly ever Sometimes Often Almost always 100% Mental Health Staffing is Sufficient* ED Leadership Supports Suicide Interventions* Suicidal Patient Treatment is a Top Clinical Priority 75% 50% 25% 0% From ED-SAFE Provider Survey (phase 1); *P under Pearson chi-square
28 Next Steps: Many Challenges Need to decide who will do suicide screening and assessment Lot of variation in the need Kaiser in CA lowest prevalence, public hospitals in CO highest Lot of variation in ED capability ED culture, provider knowledge, attitudes Very focused generally, not comfortable with mental health Size, academic, public/private, urban/rural, capitation Whoever it is will need more training! Need evidence-based ED tools for Rapid risk assessment Stratification and thresholds for hospital versus referral ED treatment Need better access to inpatient or outpatient mental health resources
29 Risk Scoring: Delirium Older ED patients, >= 1 delirium risk factors Screen 165 to detect 23 and miss 1 Han JH, et al. Acad Emerg Med Mar;16(3):
30 Discussion Where should screening occur? What does that look like? Low versus high risk? What do we need to take care of people? Training?